A system for controlling a vehicle having an autonomous mode and a semi-autonomous mode includes one or more processors and a memory in communication with the one or more processors. The memory stores a command generating module and a transmission module. The command generating module causes the one or more processors to generate, in response to an input, at least one control signal for controlling the vehicle by an envelope control system. The envelope control system utilizes a common control scheme for both the semi-autonomous mode and the autonomous mode, wherein the input is a driver input when the vehicle is in the semi-autonomous mode and the input is a pseudo-driver input when the vehicle is in the autonomous mode. The transmission module causes the one or more processors to transmit the at least one control signal to a vehicle motion controller, wherein the vehicle motion controller controls the movement of the vehicle.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for controlling a vehicle having an autonomous mode and a semi-autonomous mode, the method comprising the steps of: generating, in response to an input, at least one control signal for controlling the vehicle by a control system, wherein the control system utilizes a common control scheme for controlling the movement of the vehicle in both the semi-autonomous mode and the autonomous mode, wherein the input to the common control scheme is a driver input from a driver when the vehicle is in the semi-autonomous mode and the input to the common control scheme is a pseudo-driver input when the vehicle is in the autonomous mode; wherein algorithms utilized by the common control scheme to generate the at least one control signal remain the same for both the semi-autonomous mode and the autonomous mode; transmitting the at least one control signal to a vehicle motion controller, wherein the vehicle motion controller controls the movement of the vehicle; and wherein the semi-autonomous mode requires the driver to provide input to the control system to perform a maneuvering of the vehicle and the autonomous mode requires no input from the driver to the control system to perform the maneuvering of the vehicle.
Vehicle control systems and methods for managing autonomous and semi-autonomous driving. The invention addresses the challenge of providing consistent vehicle control logic across different levels of driver intervention. A control system generates at least one control signal to manage vehicle movement. This control system employs a common control scheme that is identical in its algorithms for both semi-autonomous and fully autonomous modes. The key difference lies in the input provided to this common scheme. In semi-autonomous mode, the input originates from direct driver input. Conversely, in autonomous mode, the input is a pseudo-driver input, generated by the system itself. This control signal is then sent to a vehicle motion controller, which executes the actual movement of the vehicle. The semi-autonomous mode necessitates driver input for maneuvers, while the autonomous mode operates without requiring any driver input for such maneuvers.
2. The method of claim 1 , wherein the pseudo-driver input and the driver input utilize a common format.
A system and method for vehicle control involves generating a pseudo-driver input to simulate human driving behavior, which is then combined with an actual driver input to control the vehicle. The pseudo-driver input is derived from a trained machine learning model that predicts optimal control actions based on vehicle and environmental data. The actual driver input is obtained from a human driver's actions, such as steering, acceleration, or braking. Both the pseudo-driver input and the driver input are formatted in a common structure, allowing them to be seamlessly integrated. The combined input is then used to adjust the vehicle's control systems, such as steering, throttle, or braking, to achieve safe and efficient operation. This approach enhances driving safety by supplementing human control with AI-driven assistance, particularly in challenging scenarios where the driver may need support. The common format ensures compatibility between the AI-generated and human inputs, enabling smooth and reliable integration. The system may also include additional features, such as real-time adjustments based on sensor data or adaptive learning to improve the AI model over time. This technology is applicable to autonomous and semi-autonomous vehicles, where human-AI collaboration is critical for safe and effective operation.
3. The method of claim 1 , wherein the driver input includes at least one of a driver command and a predicted driver command.
A system and method for vehicle control processes driver input to determine vehicle operation. The technology addresses the challenge of accurately interpreting driver intentions to enhance vehicle responsiveness and safety. The method involves receiving driver input, which may include explicit driver commands or predicted driver commands based on anticipated driver actions. The system analyzes this input to generate control signals for vehicle systems, such as steering, braking, or acceleration. Predicted driver commands are derived from historical data, driver behavior patterns, or environmental context to anticipate and preemptively adjust vehicle responses. The method ensures seamless integration between manual and automated driving modes by dynamically adapting to real-time and predictive inputs. This approach improves vehicle handling, reduces reaction times, and enhances overall driving safety by aligning vehicle actions with both current and anticipated driver intentions. The system may also incorporate sensor data, vehicle dynamics, and external factors to refine command interpretation and execution.
4. The method of claim 3 , wherein the driver command and the predicted driver command include a trajectory of the vehicle.
The invention relates to vehicle control systems, specifically methods for predicting and executing driver commands to improve vehicle maneuvering. The problem addressed is the need for accurate and responsive vehicle control that aligns with driver intentions, particularly in scenarios where real-time adjustments are necessary. The method involves generating a predicted driver command based on sensor data, such as vehicle speed, steering angle, and environmental conditions. This predicted command is then compared to an actual driver command to determine discrepancies. If a significant difference exists, the system adjusts the vehicle's control parameters to ensure smooth and safe operation. The method also includes generating a trajectory for the vehicle, which defines the path the vehicle should follow. Both the actual and predicted driver commands incorporate this trajectory, allowing the system to anticipate and respond to the driver's intended path. By continuously monitoring and adjusting based on these commands, the system enhances vehicle stability and reduces the risk of accidents. The invention is particularly useful in autonomous or semi-autonomous driving systems where precise control is critical.
5. The method of claim 1 , wherein the pseudo-driver input is a stabilizing prediction steering command.
A system and method for vehicle control involves generating a pseudo-driver input to assist in stabilizing vehicle motion. The pseudo-driver input is derived from a stabilizing prediction steering command, which is calculated based on vehicle dynamics and sensor data to counteract instability. The system monitors vehicle parameters such as yaw rate, lateral acceleration, and wheel speed to detect potential instability. When instability is detected, the stabilizing prediction steering command is generated to adjust the vehicle's steering angle, reducing or eliminating unstable motion. The pseudo-driver input can be combined with or override driver inputs to ensure safe operation. This approach enhances vehicle stability in adverse conditions, such as slippery roads or sudden maneuvers, by providing automated corrective steering adjustments. The system may also incorporate predictive models to anticipate instability before it occurs, allowing for proactive stabilization. The method ensures that the vehicle remains within safe operating limits while maintaining driver control when conditions permit.
6. The method of claim 5 , wherein the stabilizing prediction steering command is a stable lane keeping feedforward command.
A method for vehicle control involves generating a stabilizing prediction steering command to improve lane-keeping performance. The system predicts the vehicle's future trajectory and generates a feedforward steering command to maintain the vehicle within the lane. This command is designed to be stable, meaning it minimizes oscillations or deviations from the intended path. The method integrates predictive modeling to anticipate lane boundaries and road conditions, ensuring smooth and accurate lane-keeping. By using feedforward control, the system proactively adjusts steering inputs based on predicted deviations rather than reacting to them, enhancing stability and reducing driver workload. The approach is particularly useful in autonomous or semi-autonomous driving systems where precise lane control is critical for safety and efficiency. The stabilizing prediction steering command works in conjunction with other control mechanisms to provide a comprehensive solution for maintaining vehicle stability and lane discipline.
7. The method of claim 6 , wherein the stable lane keeping feedforward command includes a trajectory that substantially follows an optimal trajectory or a lane center line, wherein the lane center line is a calculated line that is located between a right edge and a left edge of a lane the vehicle is travelling on.
This invention relates to autonomous vehicle control systems, specifically methods for improving lane-keeping performance. The problem addressed is the need for precise trajectory tracking to enhance vehicle stability and safety while navigating lanes. The method involves generating a stable lane-keeping feedforward command that guides the vehicle along an optimal trajectory or a calculated lane center line. The lane center line is determined by analyzing the lane's right and left edges to define a central path between them. This approach ensures the vehicle maintains a consistent and predictable path, reducing deviations and improving overall lane-keeping accuracy. The feedforward command is derived from real-time lane detection and processing, allowing the vehicle to adapt to varying lane conditions. By following the optimal trajectory or lane center line, the system minimizes lateral deviations, enhances passenger comfort, and reduces the risk of collisions. The method integrates with existing autonomous driving systems to provide smoother and more reliable lane-following behavior.
8. The method of claim 1 , wherein: when the input is the driver input, the input is provided to a cost function to set a driver tracking cost; and when the input is the pseudo-driver input, the input is provided to the cost function to set a pseudo-driver tracking cost.
This invention relates to systems for tracking and differentiating between a human driver and a pseudo-driver in autonomous vehicle control. The problem addressed is the need to accurately distinguish between human driver inputs and automated or simulated driver inputs to ensure proper vehicle operation and safety. The invention provides a method that processes input data to determine whether it originates from a human driver or a pseudo-driver, such as an automated system or simulation. When the input is identified as a driver input, it is used to adjust a cost function that governs driver tracking, influencing how the vehicle responds to human commands. Conversely, when the input is recognized as a pseudo-driver input, it is used to modify a different cost function specifically for pseudo-driver tracking, ensuring the vehicle adapts appropriately to automated or simulated control. This differentiation allows the system to optimize tracking behavior based on the source of the input, improving safety and performance in both human-driven and autonomous modes. The method ensures seamless transitions between control modes while maintaining accurate tracking of the intended driver or system.
9. A system for controlling a vehicle having an autonomous mode and a semi-autonomous mode, the system comprising: one or more processors; a memory in communication with the one or more processors, the memory storing: a command generating module that when executed by the one or more processors cause the one or more processors to generate, in response to an input, at least one control signal for controlling the vehicle by a control system, wherein the control system utilizes a common control scheme for controlling the movement of the vehicle in both the semi-autonomous mode and the autonomous mode, wherein the input to the common control scheme is a driver input from a driver when the vehicle is in the semi-autonomous mode and the input to the common control scheme is a pseudo-driver input when the vehicle is in the autonomous mode; wherein algorithms utilized by the common control scheme to generate the at least one control signal remain the same for both the semi-autonomous mode and the autonomous mode; a transmission module that when executed by the one or more processors cause the one or more processors to transmit the at least one control signal to a vehicle motion controller, wherein the vehicle motion controller controls the movement of the vehicle; and wherein the semi-autonomous mode requires the driver to provide input to the control system to perform a maneuvering of the vehicle and the autonomous mode requires no input from the driver to the control system to perform the maneuvering of the vehicle.
The system controls a vehicle capable of operating in both autonomous and semi-autonomous modes. The vehicle uses a unified control scheme that processes inputs differently depending on the mode. In semi-autonomous mode, the system accepts direct driver inputs, while in autonomous mode, it uses pseudo-driver inputs generated by the system itself. The same algorithms and control logic are applied in both modes, ensuring consistent vehicle behavior. The system generates control signals based on these inputs and transmits them to a vehicle motion controller, which executes the necessary actions to maneuver the vehicle. In semi-autonomous mode, the driver must actively provide inputs to control the vehicle, whereas in autonomous mode, the system operates independently without requiring driver intervention. This approach simplifies the control architecture by reusing the same algorithms for both operational modes, reducing complexity and improving reliability. The system ensures seamless transitions between modes while maintaining consistent vehicle performance.
10. The system of claim 9 , wherein the pseudo-driver input and the driver input utilize a common format.
A system for vehicle control integrates pseudo-driver input and driver input, where both inputs share a common data format. The pseudo-driver input is generated by an autonomous control module, simulating human driving actions, while the driver input originates from a human operator. The system processes these inputs to control vehicle functions such as steering, acceleration, and braking. The common format ensures seamless integration and compatibility between the autonomous and human-driven control signals, allowing smooth transitions between modes. This design enables hybrid vehicle operation, where the system can switch between autonomous and manual control without requiring separate processing pathways for each input type. The unified format reduces complexity and improves reliability by standardizing data handling across different control sources. The system may also include safety mechanisms to validate and prioritize inputs, ensuring safe operation in both autonomous and manual modes. This approach addresses challenges in integrating human and autonomous control systems, particularly in scenarios requiring rapid mode switching or shared control authority.
11. The system of claim 9 , wherein the driver input includes at least one of a driver command and a predicted driver command.
A system for vehicle control integrates driver input to enhance autonomous driving capabilities. The system processes real-time driver commands, such as steering, acceleration, or braking inputs, and also incorporates predicted driver commands based on anticipated actions. These inputs are used to adjust the vehicle's autonomous control algorithms, ensuring seamless transitions between manual and autonomous driving modes. The system analyzes driver behavior to predict future commands, improving responsiveness and safety. By combining actual and predicted inputs, the system optimizes decision-making, reducing latency and enhancing overall driving performance. This approach addresses challenges in autonomous vehicle operation, particularly in scenarios requiring rapid mode switching or driver intervention. The system dynamically adapts to varying driving conditions, ensuring smooth and efficient vehicle control.
12. The system of claim 11 , wherein the driver command and the predicted driver command include a trajectory of the vehicle.
A system for vehicle control and driver assistance monitors and predicts driver commands to enhance vehicle operation. The system compares a driver's actual command with a predicted command generated by an autonomous or semi-autonomous control module. The comparison identifies discrepancies between the driver's intent and the system's predictions, allowing for corrective actions or adjustments to improve safety and performance. The system may use sensor data, vehicle dynamics, and environmental inputs to generate these predictions. In some implementations, the system integrates with adaptive cruise control, lane-keeping, or collision avoidance systems to provide seamless assistance. The driver command and predicted command include a trajectory of the vehicle, meaning the system evaluates the path or route the vehicle is expected to follow based on both the driver's actions and the system's calculations. This trajectory-based approach helps ensure alignment between human and machine decision-making, reducing the risk of conflicts or errors. The system may also adjust its predictions in real-time based on changing conditions, such as road curvature, traffic, or driver behavior. By continuously refining its predictions, the system enhances situational awareness and responsiveness, making it suitable for advanced driver-assistance systems (ADAS) and autonomous driving applications.
13. The system of claim 9 , wherein the pseudo-driver input is a stabilizing prediction steering command.
A system for vehicle control includes a pseudo-driver module that generates stabilizing prediction steering commands to improve vehicle handling. The pseudo-driver module operates by receiving sensor data from the vehicle, such as speed, yaw rate, and lateral acceleration, and processes this data to predict and stabilize the vehicle's trajectory. The stabilizing prediction steering command is a control input designed to counteract deviations from a desired path, enhancing stability during maneuvers or adverse conditions. This system integrates with the vehicle's existing control architecture, where the pseudo-driver module works alongside other control systems to provide real-time adjustments. The stabilizing prediction steering command is calculated based on predictive models that anticipate vehicle dynamics, ensuring smoother and safer operation. This approach reduces reliance on human intervention and improves automated or semi-automated driving systems by proactively stabilizing the vehicle. The system is particularly useful in scenarios where sudden corrections are needed, such as avoiding obstacles or maintaining lane discipline. By generating precise steering commands, the system enhances overall vehicle stability and control performance.
14. The system of claim 13 , wherein the stabilizing prediction steering command is a stable lane keeping feedforward command.
A system for vehicle control includes a predictive steering mechanism that generates stabilizing prediction steering commands to maintain vehicle stability and lane position. The system uses sensor data, such as vehicle speed, yaw rate, and lateral acceleration, to predict future vehicle states and generate corrective steering inputs. These inputs are designed to counteract disturbances, such as crosswinds or road irregularities, ensuring the vehicle follows a desired trajectory. The stabilizing prediction steering command is specifically a stable lane-keeping feedforward command, meaning it proactively adjusts steering based on predicted deviations rather than reactive feedback. This feedforward approach improves responsiveness and reduces the risk of overcorrection. The system may integrate with existing vehicle control modules, such as electronic stability control (ESC) or lane-keeping assist (LKA), to enhance overall vehicle handling and safety. The predictive model accounts for dynamic conditions, including varying road surfaces and environmental factors, to provide precise and timely steering adjustments. The goal is to minimize driver intervention while maintaining vehicle stability and lane discipline.
15. The system of claim 14 , wherein the stable lane keeping feedforward command includes a trajectory that substantially follows a lane center line, wherein the lane center line is a calculated line that is located between a right edge and a left edge of a lane the vehicle is travelling on.
This invention relates to autonomous vehicle control systems, specifically for improving lane-keeping performance. The system addresses the challenge of maintaining precise lane positioning by generating a stable feedforward command that guides the vehicle along a calculated lane center line. The lane center line is determined as a line equidistant between the detected right and left edges of the lane, ensuring the vehicle follows the optimal path within the lane boundaries. This feedforward command is part of a broader control system that integrates sensor data, such as from cameras or LiDAR, to detect lane markings and compute the lane center line in real time. The system dynamically adjusts the vehicle's trajectory to minimize deviations from the center line, enhancing stability and safety. By focusing on the lane center line as a reference, the system reduces the risk of lane drift caused by external disturbances or imperfect lane markings. The invention is particularly useful in autonomous driving applications where consistent lane adherence is critical for navigation and collision avoidance. The stable feedforward command ensures smooth and predictable vehicle movement, improving overall driving performance in various road conditions.
16. The system of claim 9 , wherein the command generating module further includes instructions that when executed by the one or more processors cause the one or more processors to: when the input is the driver input, provide the input to a cost function to set a driver tracking cost; and when the input is the pseudo-driver input, provide the input to a constraint function to set a constraint.
This invention relates to autonomous vehicle systems, specifically improving decision-making by integrating driver behavior and pseudo-driver inputs. The system addresses the challenge of balancing real-time driver actions with predefined constraints to enhance safety and efficiency in autonomous driving. The core system includes a command generating module that processes both driver inputs and pseudo-driver inputs, which are synthetic or simulated inputs representing ideal or constrained driving behaviors. The module evaluates these inputs differently: driver inputs are assessed using a cost function to determine the impact on tracking performance, while pseudo-driver inputs are processed through a constraint function to enforce operational limits. This dual approach allows the system to adapt to human driver behavior while ensuring compliance with safety and operational boundaries. The system dynamically adjusts control commands based on these evaluations, optimizing vehicle responses in real-world scenarios. The integration of cost and constraint functions enables a flexible yet controlled decision-making process, improving the reliability and safety of autonomous driving systems.
17. A non-transitory computer-readable medium for controlling a vehicle having an autonomous mode and a semi-autonomous mode and including instructions that when executed by one or more processors cause the one or more processors to: generate, in response to an input, at least one control signal for controlling the vehicle by a control system, wherein the control system utilizes a common control scheme for controlling the movement of the vehicle in both the semi-autonomous mode and the autonomous mode, wherein the input to the common control scheme is a driver input from a driver when the vehicle is in the semi-autonomous mode and the input to the common control scheme is a pseudo-driver input when the vehicle is in the autonomous mode; wherein algorithms utilized by the common control scheme to generate the at least one control signal remain the same for both the semi-autonomous mode and the autonomous mode; transmit the at least one control signal to a vehicle motion controller, wherein the vehicle motion controller controls the movement of the vehicle; and wherein the semi-autonomous mode requires the driver to provide input to the control system to perform a maneuvering of the vehicle and the autonomous mode requires no input from the driver to the control system to perform the maneuvering of the vehicle.
This invention relates to a vehicle control system that operates in both semi-autonomous and fully autonomous modes using a unified control scheme. The system addresses the challenge of maintaining consistent vehicle behavior across different operational modes by employing a common control algorithm. In semi-autonomous mode, the system processes direct driver inputs, while in autonomous mode, it uses pseudo-driver inputs generated by the vehicle's autonomous driving system. The same control algorithms are applied in both modes, ensuring predictable vehicle responses regardless of the operational state. The system generates control signals based on these inputs and transmits them to a vehicle motion controller, which executes the necessary actions to maneuver the vehicle. In semi-autonomous mode, the driver must actively provide inputs to control the vehicle, whereas in autonomous mode, the vehicle operates independently without requiring driver intervention. This approach simplifies system design by eliminating the need for separate control schemes for each mode, improving reliability and reducing complexity. The invention is implemented via a non-transitory computer-readable medium containing executable instructions for the control system.
18. The non-transitory computer-readable medium of claim 17 , wherein the pseudo-driver input and the driver input utilize a common format.
A system and method for processing driver input data in a vehicle involves capturing and analyzing driver inputs, such as steering, acceleration, and braking actions, to improve vehicle control and safety. The system includes a data acquisition module that collects driver input signals from various vehicle sensors and a processing module that converts these signals into a standardized format for further analysis. The system also includes a pseudo-driver input module that generates simulated driver inputs for testing and validation purposes. Both the pseudo-driver input and the actual driver input are processed using a common data format to ensure consistency and compatibility across different vehicle systems. This standardization allows for seamless integration with vehicle control algorithms, enabling real-time adjustments to enhance driving performance and safety. The system may also include a feedback module that provides drivers with real-time feedback based on their input patterns, helping to improve driving habits and reduce errors. The use of a common format for both real and simulated inputs ensures that the system can accurately compare and validate driver behavior under various conditions, improving the overall reliability and effectiveness of the vehicle's control mechanisms.
19. The non-transitory computer-readable medium of claim 17 , wherein the driver input includes at least one of a driver command and a predicted driver command.
A system for vehicle control processes driver input to determine vehicle actions. The system includes a processor that receives driver input, which may be an explicit driver command or a predicted driver command based on anticipated driver actions. The processor analyzes this input to generate control signals for vehicle systems, such as steering, braking, or acceleration. The system also includes a memory storing instructions for the processor to execute these functions. The driver input may be derived from direct user commands or inferred from predictive models that anticipate driver behavior. The control signals adjust vehicle operations in real-time to ensure safe and efficient driving. This approach enhances vehicle responsiveness by integrating both immediate and predictive driver inputs, improving overall control accuracy and adaptability. The system is designed to work with various vehicle types, including autonomous and semi-autonomous vehicles, to optimize performance based on driver intentions.
20. The non-transitory computer-readable medium of claim 19 , wherein the driver command and the predicted driver command include a trajectory of the vehicle.
The invention relates to autonomous vehicle control systems, specifically improving the accuracy of driver command predictions to enhance vehicle safety and performance. The system addresses the challenge of accurately predicting a human driver's intended actions in real-time, which is critical for seamless handover between autonomous and manual driving modes. The invention involves a method for predicting a driver's commands by analyzing sensor data, such as steering wheel position, accelerator pedal input, and brake pedal input, to generate a predicted driver command. This predicted command is then compared to an actual driver command to assess discrepancies, which helps in refining the prediction model. The system also includes a trajectory component, where both the driver command and the predicted driver command include a vehicle trajectory, allowing the system to anticipate not just discrete actions but the full path the vehicle is expected to follow. This trajectory-based approach improves decision-making in complex driving scenarios, such as lane changes or obstacle avoidance. The invention ensures that the autonomous system can accurately interpret and respond to driver intentions, reducing the risk of miscommunication between the driver and the vehicle's control systems. The method is implemented using a non-transitory computer-readable medium, ensuring reliable and consistent performance.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
August 23, 2019
March 22, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.